%ignore bug fixed in xearley (thanks to issue #44)

lark/parsers/xearley.py

@@ -52,7 +52,6 @@ class Parser:
         # Define parser functions
         start_symbol = start_symbol or self.start_symbol
         delayed_matches = defaultdict(list)
-        match_after_ignore = set()
 
         text_line = 1
         text_column = 0
@@ -89,8 +88,7 @@ class Parser:
                 m = x.match(stream, i)
                 if m:
                     delayed_matches[m.end()] += set(to_scan)
-                    if m.end() == len(stream):
-                        match_after_ignore.update(set(column.to_reduce))
+                    delayed_matches[m.end()] += set(column.to_reduce)
 
                     # TODO add partial matches for ignore too?
                     # s = m.group(0)
@@ -142,10 +140,6 @@ class Parser:
         solutions = [n.tree for n in column.to_reduce
                      if n.rule.origin==start_symbol and n.start is column0]
 
-        if not solutions:
-            solutions = [n.tree for n in match_after_ignore
-                     if n.rule.origin==start_symbol and n.start is column0]
-
         if not solutions:
             raise ParseError('Incomplete parse: Could not find a solution to input')
         elif len(solutions) == 1:

tests/test_parser.py

@@ -962,6 +962,13 @@ def _make_parser_test(LEXER, PARSER):
             tree = parser.parse("int 1 ! This is a comment")    # A trailing ignore token can be tricky!
             self.assertEqual(tree.children, ['1'])
 
+            parser = _Lark(r"""
+                start : "a"*
+                %ignore "b"
+            """)
+            tree = parser.parse("bb")
+            self.assertEqual(tree.children, [])
+
 
         @unittest.skipIf(LEXER==None, "Scanless doesn't support regular expressions")
         def test_regex_escaping(self):